Analyzing wins and losses

[cpfsf]

http://rubechat.kfan.com/viewtopic.php?f=16&t=134548



Y Axis = Quantity
X Axis = Loss Deficit

All the credit here goes to memyworld of rubechat.

[C.lupus]

Thanks for sharing. Not to bash memyworld for his nice work but something about that graph is hard for me to read (and I read lots of graphs). I wonder if it would be more intuitive if the axes were switched? Anyway, as with most data analyses, it makes me want to delve in deeper. If only I had the time...

[Yes We Kahn]

I think that graph is a little misleading. It makes it look like a majority of their games are lost by more than 10 points. However, 18 have been by less than ten and 11 have been lost by more than ten. If you are going to have a data point for 1, 2, 3, etc... you might as well continue with that as you go past 10. I'd prefer 1-5 (10 losses), 6-10 (11 losses), 11-15 (1 loss), 15-20 (2 losses), then maybe >20 (5 losses) because they are getting to be big blowouts (23, 26, 31, 32, 42).

1 - 5 :: XXXXXXXXXX
6 - 10: XXXXXXXXXXX
11-15: X
16-20: XX
20 ++ : XXXXX*

*by 23 on 12/10
*26 on 11/7
*31 on 12/11
*32 on11/2
*42 on 11/3

Pardon the medieval nature of my graph. I don't know how to do the fancy web stuff.
I'm not trying to knock the guy for putting the graph together, I just think the purpose of a graph is to quickly tell a story, and this does a better job of that. The Wolves have lost a lot of games by 10 or less. Almost half of those have been by 5 or less. They have also been blown out a few times (particularly at the beginning of the year).

[shrink]

1 - 5 :: XXXXXXXXXX
6 - 10: XXXXXXXXXXX
11-15: X
16-20: XX
20 ++ : XXXXX*

*by 23 on 12/10
*26 on 11/7
*31 on 12/11
*32 on11/2
*42 on 11/3

Pardon the medieval nature of my graph.

I like it a lot. The main function of a graph is to give a representation of numerical data in a way that's easier to understand.

Would you mind putting out "wins" in .. maybe with O's instead of X's?

It shouldn't take long.

[Yes We Kahn]

I like it a lot. The main function of a graph is to give a representation of numerical data in a way that's easier to understand.

Would you mind putting out "wins" in .. maybe with O's instead of X's?

It shouldn't take long.

With only 9 wins, not long at all. Is this what you were thinking? If I knew how to upload a graph from microsoft excel, I could put them on the same graph and make it look way better.

1 - 5 :: XXXXXXXXXX
6 - 10: XXXXXXXXXXX
11-15: X
16-20: XX
20 ++ : XXXXX*

*by 23 on 12/10
*26 on 11/7
*31 on 12/11
*32 on11/2
*42 on 11/3

1 - 5 :: OO
6 - 10: OOO
11-15: OOO
16-20:
20 ++ : O*

*by 34 on 12/4

[Esohny]

It would also help if the x-axis wasn't an absolute value.

[King Rube]

We are getting better, aren't we.

[Steve_Holiday]

the colors in the original graph were good.

[cpfsf]

We are getting better, aren't we.

This doesn't really say much, but HeHateMe stated that Minnesota didn't win 9 games until January 18.

Edit: I don't even know why I posted this.

[Yes We Kahn]

the colors in the original graph were good.

You mean to say you don't like the colors in mine?

[the_bruce]

I think a better representation is +/- with under 5 minutes in the 4th quarter. Several of those 10 pt loses came from playing the foul game when the game was well within reach at the end of the 4th.